A population-based case-control study of non-Hodgkin's lymphoma (NHL) will include 2000 patients and 2000 controls in the San Francisco Bay Area. A rapid case-finding system will be used to identify new cases in all hospitals in the six-county San Francisco Bay Area during the study period. Controls will be identified by random digit dial and through Health Care Financing Administration files and will be frequency-matched to cases by age and sex. Interview topics will include medical, diet, exercise, lifestyle, family, and sexual histories and will be conducted in study subjects' homes. Blood and buccal cells will be collected and processing of biological samples will provide large amounts of RNA, DNA, serum and viable cells for laboratory analyses. The specimens will be used to test for polymorphisms in key DNA repair, folate metabolism pathway and immune response genes, to search for novel genes and to test for HIV infection. After initial tests are completed the remaining viable blood cells, serum and genetic material will be banked in the event a new virus or genetic clues emerge. The lab results will be used with the questionnaire data to assess risk factors for NHL by histologic type, grade, sex, race/ethnicity, and HIV status. Multiple logistic regression and methods for genetic case-control data in the presence of stratification will be used for data analysis. Strengths include: 1. the large number of NHL patients in the San Francisco Bay Area provides the ability to examine associations by sex, race/ethnicity and histologic subtype; 2. innovative laboratory capabilities (e.g. ability to analyze many candidate loci for genetic polymorphisms and new polymorphisms for NHL), expert laboratory faculty and facilities available for laboratory tests; 3. complete dietary assessment using an established dietary questionnaire; 4. large number of patients and controls provides the potential to evaluate previously unstudied population subgroups and interactions among exposures related to NHL risk,- S. large male homosexual population permits exploration of HIV-related hypotheses and risks for NHL; 6. rapid case-finding mechanism minimizes loss of study subjects, particularly those with high-grade NHL; 7. population-based SEER tumor registry provides complete case ascertainment; 8. data later will be pooled with the NCI intramural lymphoma project to study histologic types and race/ethnicity groups in relation to questionnaire and some laboratory parameters.